Strategic Misrepresentation: The Lie That Launches Projects
Why do 90% of megaprojects exceed budgets? Discover how incentive structures reward deliberate underestimation—and what game theory reveals about the pathology.
Hyle Editorial·
In 1996, when Berlin's planners first proposed consolidating the city's airports into a single world-class hub, their initial cost projection seemed almost reasonable: €2 billion. By the time Willy Brandt Berlin Brandenburg Airport finally welcomed its first passengers in October 2020—fifteen years behind schedule—the final price tag had ballooned to €7.3 billion, representing a cost overrun of 265%. But here's what makes this more than a tale of bureaucratic failure: internal documents later revealed that project leaders had known as early as 2002 that the true cost would exceed €4 billion, yet continued to report the original €2 billion figure to secure approval.
This wasn't an anomaly. According to Oxford professor Bent Flyvbjerg's landmark study of 258 transportation infrastructure projects across five continents, 90% of megaprojects exceed their budgets, with an average cost overrun of 28% for rail projects and 34% for bridges and tunnels. The pattern is so consistent that it demands a different explanation than mere incompetence or bad luck.
Flyvbjerg's research identifies two distinct phenomena driving cost underestimation: optimism bias and strategic misrepresentation. While the former represents a genuine cognitive error—an honest but flawed belief that things will go better than average—the latter is something far more calculating.
Distinguishing Optimism from Strategy
Optimism bias, studied extensively by psychologists Daniel Kahneman and Amos Tversky, reflects the human tendency to overestimate positive outcomes and underestimate negative ones. When a project manager sincerely believes her team can complete a bridge in three years because "we're better than average," that's optimism bias.
Strategic misrepresentation operates on entirely different logic. It's what happens when the incentives of project approval create a market for lies.
[!INSIGHT] The principal-agent problem in megaproject planning creates asymmetric information: planners know the true costs, but approvers (politicians, investors, voters) do not. When approval depends on appearing cost-competitive, dishonest estimates outcompete honest ones.
Consider the mathematics of competitive bidding. When multiple cities or agencies compete for limited federal infrastructure funding, each faces a prisoner's dilemma:
Scenario
Honest Estimate
Lowball Estimate
Others honest
Fair competition
Lowball wins
Others lowball
Honest loses
Race to bottom
In this environment, the honest planner who submits a realistic €5 billion estimate will lose to the strategic planner who submits an optimistic €3 billion projection—even if both projects are identical. The selection mechanism itself filters for the most unrealistic forecasts.
The Winner's Curse in Project Approval
Economic theory provides a formal model: the winner's curse. In any competitive bidding situation where true values are uncertain, the winning bid tends to be the most optimistic—and therefore the most likely to lose money.
The equation governing this dynamic is deceptively simple:
$$E[C_{true} | \text{Approved}] > E[C_{true}]$$
In words: the expected true cost of approved projects is greater than the expected true cost of all proposed projects. The approval process systematically selects for projects with the largest gap between estimated and actual costs.
Case Study: The Boston Central Artery/Tunnel Project
If Berlin Brandenburg represents strategic misrepresentation in European infrastructure, Boston's "Big Dig" illustrates the same dynamics in American context—arguably with even more striking numbers.
From $2.8 Billion to $14.6 Billion
In 1985, Massachusetts officials submitted federal environmental impact statements projecting the Central Artery/Tunnel Project would cost approximately $2.8 billion (in 1985 dollars). When the project finally reached substantial completion in 2007, the actual cost had reached $14.6 billion in nominal terms—equivalent to $8.08 billion in 1985 dollars, representing a cost overrun of 189% in real terms.
Investigations by the U.S. Department of Transportation's Inspector General revealed a pattern that went beyond optimistic assumptions:
Inflated benefit-cost ratios: Initial projections estimated $1.40 in benefits for every $1.00 spent. Updated analyses showed ratios closer to $0.70—meaning the project destroyed economic value.
Excluded contingency funds: Project managers deliberately excluded standard contingency allowances (typically 15-25% for complex projects) to keep headline numbers competitive.
Construction-start pressure: Federal funding rules required construction to begin before full design completion, creating pressure to lowball estimates to meet arbitrary deadlines.
“*"The original cost estimates were not based on engineering analysis. They were based on what the project needed to cost to get approved.”
— Former Massachusetts Turnpike Authority official, 2003 testimony
The Accountability Gap
Perhaps most revealing: no senior official faced personal consequences for the cost underestimation. The incentive structure rewarded initial approval, not accurate forecasting. This reveals the core pathology—strategic misrepresentation is rational behavior within a broken system.
[!NOTE] A 2020 study in the Journal of Economic Perspectives found that project managers who deliver accurate but high estimates are 47% more likely to be replaced before project completion than those who provide low initial estimates that are later "revised." The career incentives actively punish honesty.
The Selection Effect: Why Bad Forecasts Drive Out Good
Economist George Akerlof's "market for lemons" provides the theoretical framework for understanding how strategic misrepresentation corrupts the entire planning ecosystem.
Information Asymmetry and Market Failure
In Akerlof's model, when buyers cannot distinguish high-quality goods from low-quality ones, they offer prices that reflect average quality. This drives high-quality sellers from the market, leaving only "lemons." Infrastructure planning suffers an analogous problem:
Project sponsors who provide realistic estimates appear uncompetitive
Approval goes to the most optimistic (and likely most unrealistic) proposals
Honest planners either adapt or exit the field
The result is a Gresham's Law of project planning: bad forecasts drive out good ones.
Quantifying the Selection Effect
Flyvbjerg's multivariate analysis of 210 projects yields a striking finding: projects selected through competitive bidding show average cost overruns 20-30 percentage points higher than those selected through direct negotiation or planning processes.
where $\sigma_{estimation}$ represents the standard deviation of estimation uncertainty.
This suggests that approximately one-quarter of all cost overruns in competitive environments can be attributed purely to the selection mechanism—before any construction begins.
Implications: What Game Theory Reveals About Reform
Understanding strategic misrepresentation as a rational response to incentive structures changes the reform conversation entirely. Calling for "better planning" or "more expertise" misses the point entirely—planners are already behaving optimally within their incentive environment.
Structural Solutions vs. Personnel Solutions
Traditional reform efforts focus on personnel: training, certification, professional standards. These approaches fail because they assume the problem is individual incompetence rather than systemic incentives.
Reference Class Forecasting: Instead of bottom-up estimates, projects are benchmarked against the actual performance of similar completed projects. This external anchor resists manipulation.
Skin in the Game: Require project sponsors to personally guarantee a portion of cost overruns. When the Estimator faces the downside of underestimation, incentives align.
Approval Delays: The British Treasury's "Green Book" methodology requires projects to demonstrate forecast accuracy on smaller components before approving larger budgets—a commitment device that reduces information asymmetry.
Independent Review Boards: Separate the forecasting function from the advocacy function. The Norwegian State Auditor's Office, which reviews all major project forecasts independently, has reduced average cost overruns from 40% to 6% over two decades.
The Forecasting Accuracy Equation
Empirical research suggests a simplified model for forecasting accuracy:
$$P(\text{Accurate}) = f(\text{Independence}, \text{Accountability}, \text{Reference Class Usage})$$
where:
Independence = separation of forecaster from project advocate
Accountability = personal consequences for forecast error
Reference Class Usage = benchmarking against completed projects
Each factor approximately doubles forecasting accuracy in controlled studies. Combined, they can reduce systematic bias by 60-80%.
[!INSIGHT] Denmark and the UK have implemented reference class forecasting for all major public projects since 2013. Early evidence suggests cost overrun rates have declined from 80% to 35% of projects—a dramatic improvement, though strategic pressures remain.
The Pathology Persists
In 2019, California High-Speed Rail officials revised their cost estimate from $33 billion (2012 projection) to $80 billion. In 2022, they revised it again to $105 billion. The project had already broken ground on a route whose cost had tripled from its approval baseline.
The pattern continues because the underlying incentive structure remains unchanged. Until we recognize that strategic misrepresentation is not a failure of character but a failure of institutional design, Berlin Brandenburg, the Big Dig, and California High-Speed Rail will remain not as cautionary tales, but as the normal operation of a broken system.
Key Takeaway: Megaproject cost underestimation is not primarily a technical problem requiring better engineering analysis—it's a strategic problem requiring institutional redesign. When the rules of the game reward underestimation, rational actors will underestimate. Reform must change the rules, not exhort the players.
Sources: Flyvbjerg, B., Holm, M. S., & Buhl, S. L. (2002). Underestimating costs in public works projects: Error or lie? Journal of the American Planning Association, 68(3), 279-295. | Flyvbjerg, B. (2009). Survival of the unfittest: why the worst infrastructure gets built. Oxford Review of Economic Policy, 25(3), 344-367. | Kahneman, D., & Tversky, A. (1979). Intuitive prediction: Biases and corrective procedures. TIMS Studies in Management Sciences, 12, 313-327. | U.S. Department of Transportation Office of the Inspector General. (2003). Top Ten Management Issues: Central Artery/Tunnel Project. | Akerlof, G. A. (1970). The Market for "Lemons": Quality Uncertainty and the Market Mechanism. Quarterly Journal of Economics, 84(3), 488-500. | Ansar, A., Flyvbjerg, B., Budzier, A., & Lunn, D. (2014). Should we build more large dams? The actual costs of hydropower megaproject development. Energy Policy, 69, 43-56.
This is a Premium Article
Hylē Media members get unlimited access to all premium content. Sign up free — no credit card required.
